UNIVERSITY OF CALIFORNIA PUBLICATIONS. COLLEGE OF AGRICULTURE. AGRICULTURAL EXPERIMENT STATION. A New Method oi Making Dry Red Wine. By FREDERIC T. BIOLETTL BULLETIN No. 177. (Berkeley, Cal., February, 1906.) W. W. SHANNON SACRAMENTO: : : : superintendent state printing. 1906. BENJAMIN IDE WHEELER, Ph.D., LL.D., President of the University EXPERIMENT STATION STAFF. E. W. HILGARD, Ph.D., LL.D., Director and Chemist. (Absent on leave.) E- J. WICKSON, M.A., Acting Director and Horticulturist. W. A. SETCHELL, Ph.D., Botanist. ELWOOD MEAD, M.S., CE-, Irrigation Engineer. C. W. WOODWORTH, M.S., Entomologist. R. H. LOUGHRIDGE, Ph.D., Agricultural Geologist and Soil Physicist. {Soils and Alkali.). M. E- JAFFA, M.S., Assistant Chemist. (Eoods, Nutrition.) G. W. SHAW, M.A., Ph.D., Assistant Chemist. (Cereals, Oils, Beet-Sugar.) GEORGE E. COLBY, M.S., Assistant Chemist. (Fruits, Waters, Insecticides.) A. R. WARD, B.S.A., D.V.M., Veterinarian and Bacteriologist . E. W. MAJOR, B.Agr., Animal Industry. RALPH E- SMITH, B.S., Plant Pathologist . E. H. TWIGHT, B.Sc, Diplome E.A.M., Viticulturist. F. T. BIOIvETTI, M.S., Viticulturist. WARREN T. CLARKE, B.S., Assistant Entomologist and Asst. Supt. Farmers' Institutes. H. M. HALL, M.S., Assistant Botanist. GEORGE ROBERTS, M.S., Assistant Chemist, in charge of Fertilizer Control. C. M. HARING, D.V.M., Assistant Veterinarian and Bacteriologist . ALBERT M. WEST, B.S., Assistant Plant Pathologist. E- H. SMITH, M.S., Assistant Plant Pathologist. G. R. STEWART, Student Assistant in Station Laboratory . ALICE R. THOMPSON, B.S., Assistant in Soil Laboratory. D. L. BUNNELL, Clerk to the Director. R. E- MANSELL. Foreman of Central Station Grounds. JOHN TUOHY, Patron, ) > Tulare Substation, Tulare. J. FORRER, Foreman, ) J. W. MILLS, Pomona, in charge Cooperative Experi ments in Southern California . J. W. ROPER, Patron, HENRY WIGHTMAN, In cht ROY JONES, Patron, f > University Forestry Station, Santa Monica. J. H. BARBER, Foreman, ) VINCENT J. HUNTLEY, Foreman of California Poultry Experiment Station, Petaluma. \ University Forestry Station, Chieo. 'large, ) The Station publications (Reports and Bulletins), so long as avail- able, zv ill be sent to any citizen of tJie State on application. A NEW METHOD OE MAKING DRY RED WINE. In Bulletin No. 167 of this Station, discussing various means of improving the quality of dry wines made in hot countries, the conclu- sion was reached that the method which offered the best hope of attain- ing the end in view was that by which the requisite tannin and color were extracted by heat before fermentation. The method, as outlined there, is as follows: "1. Heating the crushed grapes to a temperature and for a time sufficient to extract the necessary color, tannin, and body. "2. Immediate separation of the must, and cooling to 85° F. u 3. Immediate fermentation of the must at a temperature not exceed- ing 90° F." The bulletin further states: "The present status of the method is this : " 1. It has been shown, both in California and in France, that it is possible, when working with small quantities, to attain the object in view by this method. "2. The method has been used with success in France in the whole output of a cellar manufacturing 75,000 gallons of wine in a season.'' Experiments, made on a sufficiently large scale during the past vintage, have further demonstrated that the method is perfectly adapted to large-scale practice in California, and that it is possible by this means to produce a wine of excellent quality where it had been found impracticable to do so by the old methods. These experiments indicate strongly, moreover, that it is not only possible to make good dry red wine by this means, but that, where the climate is hot during the vintage, it is the simplest and most economical method of attaining the end in view when working on a large industrial scale. This is not intended to mean that the use of this method alone is all that is necessary to improve the quality of the dry wines of our hotter regions, but that, together with proper attention to the ordinary details of wine-making, the proper use of cooling machines, and perhaps of pure yeast, we have a means by which we can be sure every year of making the best wine that the grapes available are capable of produc- ing. In a cellar properly arranged for the purpose, all this can be accomplished at a cost so little in excess of the cost of the usual methods of manufacture that the loss which occurs in practically every large 4 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. winery in the State, of a few vats of wine by bad or incomplete fermen- ation, far more than counterbalances it. The experiments show further that, where the grapes are well ripened and in fair condition, all that is necessary in our hot valleys, for the production of good, sound wine, is the use of cooling devices in the fermentation, and attention to proper sterilization of vats and casks. In order to give a clearer idea of how the method could be carried out on a large scale in a winery built for the purpose, a plan of such a winery has been given with enough detail to exemplify all parts of the process and to enable any one to calculate the cost of operation. While the main object of the experiments was to test the capabilities of the method referred to, other methods and other matters of interest in the fermentation of wine were investigated both for comparison and for the purpose of devising the best way of carrying out the method in practice. The main lines of investigation were the following: 1. Extraction of color and tannin by artificial heating. 2. Control of the temperature of fermentation by means of a cooling machine. 3. The use of pure and selected yeast. 4. The use of sulfurous acid in controlling the temperature of fer- menting grapes. 5. The making of wine from incompletely ripe grapes. In order to make the bearing of the experiments clear, a detailed account of the method used is necessary. Reference to pages 19 to 26 of Bulletin No. 167 will make the objects of the principal experiment clear. There, will be found a discussion of the effects of high tempera- tures on the fermentation of wine, and an account of the preliminary laboratory experiments which led to the devising of the method used in the larger Avinery experiments this year. It will be seen, from the reference given, that high temperatures have beneficial and necessary, as well as injurious, effects in the production of dry red wine, and the experiments detailed in this bulletin show that a practical method has been devised which will "combine the beneficial effects of heat in the extraction of color, tannin, and body with those of cool fermentation in producing bouquet, freshness, and maximum amount of alcohol." These experiments also indicate very strongly that by use of the method recommended we are certain to produce a sound wine from any grapes in fair condition, however hot the season; and that, even when the grapes are inferior by reason of mold, drying up, etc., much better wine can be made than by any of the usual methods. The principal merit of the method, therefore, is the certainty it gives that every rat of wine will be perfectly sound every year. A NEW METHOD OF MAKING DRY RED WINE. D DESCRIPTION OF METHOD. Crushing. — The grapes were taken as they came to the winery in the usual way, passed through the ordinary crusher and stemmer, and pumped by means of a must-pump through the regular must-line into 1,500-gallon vats. As the crushed grapes came into the vat a small amount of potassium meta-bisulfite dissolved in water was added gradually, in such a way as to distribute it equally throughout the vat. The amount i ised varied from .2 to .4 per mil. of the weight of the grapes, that is 1o say, from 6 to 12 ounces per ton. The sulfite was added to paralyze temporarily the action of the molds, bacteria, and yeast present, and to facilitate the solution of the color. Another important use of the sulfite is to prevent too much oxidation during the heating, and so to avoid the " rancido " taste which might be acquired if the grapes were allowed to remain hot too long or if they were heated too high. Preparation of Vats. — The vats were cleaned with hot water and soda and then swabbed with a 3% solution of commercial sulfuric acid to partially disinfect them. The acid solution was left on the walls of the vats for a few hours and then rinsed off with water. Straining. — At the bottom of each vat was placed a strainer extend- ing from the bunghole across the whole bottom to the opposite side- The form of strainer used is shown in Fig. 1. It consisted of an G FIG. 1. Strainer for bottom of Fermenting Vat. inverted rectangular trough closed at one end and having a top consist- ing of a solid board, and sides consisting of slats made with beveled edges like the slats of a press basket. These strainers were found to be not sufficiently effective, owing to the very thorough way in which the grapes were broken up by the method of handling. In practice it would be necessary to have a more efficient method of straining, or to adopt a method of crushing the grapes and conveying them to the vats, which would not result in such complete maceration. The former, i. e., the use of a more efficient straining device, would be preferable, as it is very desirable to have the grape-pulp thoroughly disintegrated in order to facilitate the extraction, which takes place the more rapidly the more completely the grapes are crushed. 6 UNIVERSITY OF CALIFORNIA— EXPERIMENT STATION. As soon as the vat was full of crushed grapes the must was allowed to run off the skins into another vat. It was found possible to run off a volume of must equal to only one half of the total volume of the crushed grapes. With a better arrangement for straining, a volume equal to two-thirds could be run off, which would much facilitate the subsequent operations. Heating. — As soon as the must was separated it was passed through the heater and back into the vat containing the strained skins. The must was heated to from 140° to 150° F., and as the grapes were much cooler thev remained at the bottom, and it was found verv difficult to mix the two when the hot must was pumped on to the top. This diffi- culty was overcome in great measure by attaching the hose, bringing the must from the heater, to a faucet in the bottom of the vat. By this means the hot must was forced through the cool grapes, and a thorough stirring at the end equalized the temperature in the whole vat. The Heater. — The must was heated by means of a " must-heater ' : constructed by Gomot, of Nimes, France. This machine is essentially a large tubular pasteurizer, the exterior of which is shown in the figure on the cover. The height of the boiler is 6 feet 10 inches from the ground to the top, and the chimney is 2 feet 5 inches more. Its diameter is 4 feet 3-j inches, and the tube connections protrude a few inches more on all sides. It consists of a system of copper tubes, through which the wine runs, and a boiler surrounding them, which is heated by a fire in the firebox below, from which the heat passes by four flues into the chimney. When used in the way intended by the manufacturer, the boiler is filled with water. In our experiments, acting on the suggestion of Mr. Meakin, Ave found it more convenient to simply admit steam into the boiler and to dispense with both water and fire. There are eighty straight copper tubes joined by unions outside the boiler, as shown in the figure. These tubes are each 3 feet 10 inches long and 1^ inches in diameter, and are easily cleaned by removing the couplings, for which purpose the arrangement is very simple. In our experiments, there was no deposit whatever in the tubes, and after pass- ing water through the machine they were found perfectly clean. The machine is furnished with a strainer, which prevents seeds and skins passing into the tubes. This strainer must be opened and cleaned occasionally, and is so constructed that this is easily done. Four ther- mometers are placed in various parts of the heater, which enable the operator to control the heating perfectly. There is one thermometer in the boiler, one at the entrance of the must, one at the exit, and one half- way between these. By carefully watching these thermometers and A NEW METHOD OF MAKING DRY RED WINE. 7 manipulating the steam valve, the heating can be regulated in a few minutes to any degree required. The heating remains very constant so long as the steam pressure does not vary, and providing the supply of must is regular. A hand pump can not be used to supply the must, as the output is too irregular. It is necessary to have a pump that will give a constant stream of uniform volume. For this purpose a steam pump would be excellent, as its output can be regulated to any desired rate. For the experiments a gasoline motor pump especially constructed by Gomot for the purpose was used. The rate of pumping could be varied by a FIG. 2. Gomot Motor Pump. simple arrangement to from 375 to 1,500 gallons per hour. Such control of the rate of pumping is necessary, as the heating depends on the rate with which the must passes through the machine, as well as upon the amount and pressure of the steam admitted. Capacity of Heater. — It was found that the machine would heat 1,000 gallons of must per hour from 75° to 150° F. when the temperature of the boiler was kept at 185° F. The steam pressure during the trials was 80 pounds. There was no pressure in the boiler of the heater, as it has an opening above. Heating the must to 150° F. is about what would be found convenient in practice if the temperature of the crushed grapes was 75° F. and two- 8 UNIVERSITY OF CALIFORNIA— EXPERIMENT STATION. thirds of their volume of must was passed through the heater at 150° F. This is just about what is necessary to give the desired results. Extraction. — When the mass of crushed grapes has reached the desired temperature, it is left, with occasional stirrings (every three or four hours), until the required color and tannin have been extracted by the must. It has been shown by the experiments detailed on page 22 of Bulletin No. 167 that the color and tannin are extracted pari passu, so that by periodical observations of the color it is possible to note the progress of extraction. Salleron's Vino-Colorimeter. — For this purpose a Salleron vino-color- imeter is extremely useful. A very full description of the nature and use of this instrument was given by Prof. E. W. Hilgard in the Viti- FIG. 3. Salleron's Vino-Colorimeter. cultural Report for the seasons 1885 and 1886, on pages 23 and 26. So far, this is the best and most practical means which has come into general use for measuring the color of wine, and as some quick and reliable means of determining the color during the process of extraction is necessary for the proper carrying out of this method, a short descrip- tion of the instrument is given here. The colors of wines differ in two respects. They may differ in tint (that is to say, the kind or nature of the color may differ), and they may differ in intensity (that is, the depth or amount of color may differ). In order, therefore, to measure the color of a wine and to compare it with the color of another wine we must measure both its tint and its intensity. Both these measurements may be made with the Salleron vino-colorimeter. For the determination of the tint a scale (Fig. 3, G-H) is furnished, consisting of a piece of cardboard on which are gummed small disks of silk which represent all the tints usually found in dry red wines. At the top of the scale is a disk marked VR (violet-red), which is the tint of a new wine made from a good coloring grape with full acidity, such as the Petite Sirah or Cabernet. At the bottom of the scale is a disk A NEW METHOD OF MAKING DRY RED WINE. 9 VINOCOLOFilttLTME ,1 a \ A i^^« i — * ; 1MES» — El — ' H BL7*«I £ B Or- 1 3H- — S^" ,, ■ ^ — B VK W V.RV VR YR ( marked 3R (third red), which is the tint of an inferior coloring grape with low acidity, such as Grenache or Mission. Between these extremes are eight other disks representing the intermediate tints which red wines may have. Altogether there are ten disks repre- senting ten tints, which are marked as indicated in Fig. 4. Each tint differs from the one above it in having a slightly greater admixture of yellow. Wines of claret type should not fall below the 5V R . while those of Burgundy type may go to the bottom of the scale. Ports contain more yellow even than the 3R, and can not be measured satisfactorily by this scale. As wines become older the tint gradually becomes yel- lower, so that a wine which when made corresponds to the VR disk may fall to the 5VR disk when it is a year old. All the disks are made with the same intensity of color, «o that in order to compare wines with them it is necessary to re- duce the inten- sity of color of the wines to that of the disks. This is done by varying the thickness of wine throng h which we look. If we observe the color of a wine by looking through it in a glass three inches in diameter it will appear to be twice as dark or intense as if w r e look through it in a glass of only one and one-half inches diam- eter. This principle is made use of, both to bring the wine to the standard intensity which makes it possible to compare its tint, and also to measure the FIG. 5. Construction of Salleron's Vino-Colorimeter. ^ To use the instrument a small portion of the wine is placed in the receptacle V (Fig. 5), and the screw cap ab moved up or down until the thickness of wine between the glass R v J.SALIfRON FIG. 4. Scale of Tints. 10 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. « disks c and d through which we look gives us an intensity equal to that of the disks. The scale is then moved back and fortli until the proper tint is found. We now have all the data for describing the exact color of the wine. The tint is given by the name of the disk used and the intensity by the distance of d from c, which is read off from the scale on the outside of the eap ab at the point marked on AB with an arrow. If we get such a reading as 3VR, 160 it means that the tint of the wine is third violet red and the standard intensity is obtained by looking througli 160 hundredths of a millimeter of the wine. If we get the reading 80 on the scale it will show that only half as much wine is necessary to give the standard intensity and, therefore, that the wine is twice as dark or intensely colored as in the first case. The smaller the reading on the scale, therefore, the more deeply colored the wine. To overcome this inconvenience Professor Hilgard has adopted a scale of which the notation is inversely proportionate to the thickness of the wine necessary to give the standard tint, and of which, therefore, the numbers given are directly proportionate to the intensity of color. For this purpose he has taken a wine which required 40 hundredths of a millimeter to give the standard intensity as an arbitrary standard, which he calls 100. A wine with half this amount of color would then, with this scale, have an intensity of 50. The number is obtained by dividing 4,000 by the reading of the scale. This would give us, then, for the first example an intensity of 4000~:-160, or 25, and for the second 4000^80, or 50. If R represents the reading on the scale, then 4000 -f-R = Intensity by Professor Hilgard's scale. Comparison of Scales. Salleron Scale. Hilgard Scale. 40 = 100 80 50 160 = 25, etc. The Hilgard scale has the great advantage of giving a clear idea of the intensity of color in a wine without the need of making any mental calculations, for if the color, of one wine is represented by a number three times as large as that representing the color of another we know that the first has three times as much color as the latter. Professor Hilgard's notation has been used in the account of experi- ment work detailed later. In the absence of any recognized standard of intensity for the color of wines it is difficult to say what the proper degree of color is. In order to give some idea of what the numbers of Professor Hilgard's scale indicate in practice, it may be said that a wine measuring 15 in A NEW METHOD OP MAKING DRY RED WINE. 11 intensity would be considered sufficiently dark for an ordinary table wine. Any wine which falls below 10 is too light in color, while any- thing over 20 may be considered darker than is necessary. These figures have reference to the United States. In Europe the popular taste does not require quite so much color. A young wine should contain more color than these figures indicate, as a certain proportion of the color always drops during aging. This loss of color is most rapid during the first two months after fermenta- tion. It becomes less rapid later, until after four months the loss is very slow. The amount of loss varies very much with different wines, and with different methods of handling, but a Zinfandel to have 15 of color when it is two years old should have at least 35 or 40 directly after drawing off from the fermenting vat, and, to preserve its color even so well as this, it must have full acidity and must not be aged too quickly by frequent racking or keeping in small casks or warm cellars. The following table is appended to show the actual loss of color found by observation. The tests were made with experiment wines fermented at the Station cellar, and were all made in small quantities and kept in casks of from 10 to 20 gallons. In larger quantities the loss of color would be slower, and the depth, of color indicated for four months probably corresponds to what would be found in practice in a wine of the same character at the end of eighteen months or two years: TABLE I. Loss of Color during Aging of Wine. Number of Wines Color. Examined. At Pressing 1 Month. 2 Months. 3 Months. 4 Months. Four ___ _. . 330 140 85 35 23 13 Loss. 191 42.1% 95 32.2% 50 41.2% 25 28.6% 15 34.8% 9 30.7% Loss. 117 64.5% 91 35.0% 36 57.6% 16 54.3% 11 52.2% 9 30.7% 114 68 34 15 10 7 Loss. 65.4% 51.4% 60.0% 57.1% 56.5% 46.2% 101 59 27 13 9 5 Loss. 69.4% 56.8% 68.2% 62.9% 60.9% 61.5% Six Five Seven .. _ _'_ Five Three Average of color remaining 100 65.1% 50.9% 14.2% 43.9% 36.5% Average loss of original color for each month . 34.9% 7-0% 7-4% It will be seen by examining the table that approximately the same proportion of the color is lost whatever the original color in the wine, so that it is possible to foretell within very close limits what the color of a wine will be at the end of a certain time when kept under certain conditions. This applies only to wines of which the coloring matter is normal. Wines made from such grapes as Lenoir, Grenache, Trous- seau, or from partially dried or moldy grapes, lose their color more quickly. 12 UNIVERSITY OF CALIFORNIA— EXPERIMENT STATION. It may be said, then, that wines kept in very small casks lose about one-third of their color during the first month, about one-half during the first two months, and nearly two-thirds of their color during the first four months after pressing. The loss at first is much more rapid in the very deeply colored wines than in those more lightly colored, as is shown by the curves in Fig. 6. FIG. 6. Loss of Color in Red Wines during the First Four Months. The numbers on the right show the color at pressing; those on the left at four months. The dotted curve shows the average loss calculated from observations on thirty wines. Cooling. — As soon as the extraction of the skins has proceeded far enough, as indicated by the reading of the colorimeter, the must, now containing tannin and color, is drawn off and cooled to 80° or 85° F. This was done in the experiments by means of the cooling machine described in Bulletin No. 174. As the must is hot (125° F.) and the temperature has to be lowered only to 80° or 85° F., the cooling takes place very easily and rapidly with the cooler used. The hot must runs off very freely, as the pulp cells of the grapes have all been thoroughly A NEW METHOD OP MAKING DRY RED WINE. 13 heated and have lost the power of retaining the juice possessed^by the cells of fresh grapes. A certain amount of juice remains, however, in the skin, most of which can be recovered by pressing, as is ordinarily done with fermented grapes. It would be better to extract this juice by some diffusion process, but so far no satisfactory method of doing this has been devised. In the experiments, only the must which ran off without pressing was used, but it would be very desirable to obtain that remaining in the skins, as it contains more color than that in the free run. A continuous press would be excellent for this purpose, as the maceration of the pomace which occurs with such presses would not have the bad effect it has when pressing fermented pomace, because the extra amount of solid matter would be to a great extent precipitated during fermentation. As soon as the must is cooled and run into fermenting vats, any correction which it needs, such as the addition of acid or water, should be made. At the same time a starter of yeast should be used, preferably of tested pure yeast. Pure Yeast. — In the experiments, a pure Champagne yeast was used. This yeast had been thoroughly tested and found capable of fermenting out very sweet musts and producing a good wine. The same yeast was used in all the experiments and was found to give good results with both white and red wines. Method of Using the Pure Yeast. — A method was devised for keeping a supply of pure yeast on hand for the various fermentations, which was found very simple and convenient. As the method is of general application and could be used for the same purpose with the ordinary methods of wine-making, it is described here. The method adopted was as follows: Starting with a 4-ounce flask, the pure yeast was increased by pouring it into a large flask containing one gallon of must which had just been cooled to 90° F. after sterilizing by boiling. Three days later, this gallon of must was fermenting well and contained a large amount of active yeast. In the meanwhile 25 gallons of must, obtained from clean, sound grapes, had been cleared of its sedi- ment and yeast by settling with sulfurous acid. This was done by adding 1 ounce of potassium meta-bisulfite dissolved in one pint of water to 25 gallons of must. This had the effect of preventing fer- mentation until all sediment, including the molds and yeast, in the must had settled. In twenty-four hours, must treated in this way was perfectly bright and was racked off into a clean cask which had been thoroughly sterilized by steaming. Must treated in this way is prac- tically free from all fermentative organisms. Five gallons of this clear must were then freed from the sulfurous acid by boiling and placed in a shallow tub which had been thoroughly cleaned and sterilized with 14 UNIVERSITY OF CALIFORNIA— EXPERIMENT STATION. boiling water. As soon as the must in the tub had cooled to 90° F., the gallon of yeast in the flask was added and the whole thoroughly aerated by dipping up the must with a gallon measure and pouring it through the air back into the tub. As soon as fermentation was well under way (twenty-four hours), 2-J gallons of the clear must were added without previous boiling. More than this should not be added for fear of the arrest of the fermentation which might occur if too much of the must containing sulfite were added at once. The next day, 5 gallons more of the clear sulfited-must were added. It was safe to add more this time as the yeast was becoming used to the sulfurous acid. Twenty-four hours later the remaining 12-J gallons of the sulfited-must were added. In this way, at the end of six days, 25 gallons of must containing pure yeast w r ere obtained practically free from molds, bacteria, or wild yeast. In order to have a maximum amount of vigorous yeast in the must it should be thoroughly aerated several times every day, and fermented in a low, wide tub. As soon as a stock of pure yeast is obtained in this way it can be kept up simply by replacing the yeast taken out with an equal quantity of clear sulfited-must prepared in the way described. During the vintage it is not necessary to crush and press grapes specially for this purpose, but the must can be taken directly from the vats or casks, care being taken to sulfite the must as soon as it is separated from the grapes before the slightest fermentation has commenced. A stock of 25 gallons is sufficient for the fermentation of 2,000 gal- lons per day. Twenty gallons may be taken from the pure yeast tub every day and replaced with sulfited-must, and if the temperature is kept above 80° F. and the must is not too sweet (not over 22% B.) and sufficient aeration is given, the amount of yeast in the must will not diminish. Every care should be given to prevent contamination of the yeast. The yeast tub should be kept covered with a clean cloth and should be placed in a room separate from the fermenting vats. All tubs, buckets, casks, hose, etc., used in the production of the yeast should be thor- oughly sterilized with boiling water and not used for other purposes. While it is not pretended that by these means, however carefully carried out, we are sure to obtain an absolutely pure culture of yeast, in the sense that it does not contain a single germ or cell of anything but the yeast we start with, it was found that, for practical purposes, it preserves the culture pure during the whole vintage. As we add the yeast to grapes containing spores and germs of many kinds or to imper- fectly sterilized must, the few contaminating spores that may get into our pure yeast are quite harmless. This is true, however, only if we exercise well the precautions indicated. If we allow vinegar flies to get at our yeast vats, or draw off the yeast with hoses used for racking wine or must, the yeast will quickly become badly contaminated. A NEW METHOD OF MAKING DRY RED WINE. 15 Fermentation, — After the red must has been cooled and the yeast added, fermentation will start immediately. Within twenty-four hours the temperature will rise to 90° or 95° F., depending on the outside temperature and the size of the vat. Before it reaches 95° F. the fer- menting must is passed through the cooler and reduced to 80° F. If the weather is not very hot, the vats do not contain more than from 1,000 to 3,000 gallons, and the original density of the must does not exceed 22% to 23% Balling, this cooling will be sufficient. Usually, however, if we are to have a cool fermentation, another cooling later will be necessary. In the experiments two coolings were always given, and wines containing over 14% of alcohol and in one case over 15% fermented out perfectly dry. After Fermentation. — With musts containing no more than 22% Bal- ling, it is possible to obtain perfectly dry wines in four to five days in the fermenting vats. To insure this, a certain amount of aeration should be given at each cooling. This is accomplished simply by allow- ing the stream of cooled must to fall through the air a distance of three or four feet into the vat. The must will in this way carry down enough air to keep the yeast working vigorously. This aeration will at the same time get rid of any excess of sulfurous acid which may be present. In the experiment wines, this was accomplished so perfectly that only the faintest traces of the acid were found in the finished wines, an amount less than one-fiftieth of that allowed in French wines imported into the United States. More aeration than is necessary for the completion of the fermenta- tion should be avoided, on account of its effect on the color, which is diminished in quantity and injured in quality by excessive exposure to the oxygen of the air. With musts containing over 24% Balling it will usually be found impossible to ferment the wine perfectly dry in the four days, but it is inadvisable to leave it in the open fermenting vats any longer than this. At the end of four days, whatever the strength of the original must, the fermentation will be very slow. If the temperature, however, has never exceeded 95° F. it will not have ceased entirely, and every effort should be made to keep it going until the wine is perfectly dry. It is a capital mistake to allow the fermentation to stop and the yeast to settle while the wine still contains .5%, 1%, or even 2% of sugar, as is often done in the hope that the fermentation will recommence in the spring and eliminate this sugar. At the end of four days in the open vat the wine should be trans- ferred to a storage cask however much sugar it still contains. This transfer should be made with thorough aeration, and the receiving cask should not be sulfured. The wine should run in a stream which falls 16 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. through the air into the pump-tub, or well, and should enter the storage cask through the upper bunghole so that it will fall through the air in the cask, unless some other means is adopted for supplying the needed oxygen. This aeration will reinvigorate the yeast and will usually be sufficient to keep the fermentation going until the wine is dry. The wine should be watched, however, and, if in seven days it is not dry, it should be aerated again by running it off from the bottom of the cask and pumping it back into the same cask. In the experiments this was found sufficient, even in cases where the resulting wine showed over 15% of alcohol, to bring the wine to dryness in three weeks from the time the grapes were crushed. By this means the dangerous bacterial fermentations, which so often injure wines, even when they do not spoil them, during the one or two months following the vintage, will be avoided. The presence of a small quantity of sugar in the wine can not be detected by means of the ordinary saccharometer or mustimeter used by cellermen. The saccha- rometer may descend to the mark, or even considerably below in wines which contain over 1% of sugar. This is especially true of highly alco- holic wines, which are the most troublesome to get dry. The taste of the wine-maker is the only practical means of determining the presence of a dangerous residue of sugar in the wine. So long as a taste of sweetness can be perceived in the wine by a practiced taster, means should be adopted to keep the fermentation going. For this purpose, usually all that is necessary is an occasional aeration and stirring up of the yeast, as already described. Prompt action, however, is necessary. The wine should never be allowed to get cold before it is quite dry. If the wine is placed in puncheons or other small casks it is very useful, if not quite necessary, to keep it in a room where the temperature does not fall below 70° F. If it is placed in large casks (1,000 to 5,000 gallons or larger), it will retain its heat for a sufficient time to become perfectly dry if the needed aeration is given. If, even with these measures properly carried out, the wine still remains sweet, it means, provided the grapes used were not excessively moldy, that the original must contained more sugar than it is possible for the yeast to eliminate. If the must contains more than 28% Ball- ing, as will sometimes happen, especially when partially dried grapes are present, there is usually no possibility of fermenting it dry. In this case, the sugar may be eliminated by blending with a wine contain- ing less than 12% of alcohol. This blending, however, must be done promptly, before either w T ine has become cold and before the wine con- taining sugar has become injured by bacterial fermentation. If a wine is not perfectly dry within seven days after being transferred from the fermenting vat to the storage cask, an alcohol determination should be A NEW METHOD OF MAKING DRY RED WINE. 17 made. If there is more than a mere trace of sugar and the alcohol exceeds 14-J%, the wine should be blended immediately with a weaker wine and aerated again. First Racking. — While, so long as any sugar remains, our efforts should be directed toward keeping the yeast vigorous and suspended in the wine, we should use means to accomplish the opposite results as soon as the sugar has all disappeared and the wine is perfectly dry. The function of the yeast is to change the sugar of the must into the alcohol of the wine. As soon as this is effected the wine should be freed from the yeast as soon as possible. If the fermentation has been conducted successfully, the yeast will commence to deposit immediately and the wine will be comparatively clear within two or three weeks after it has become dry. The first racking, then, should take place at this time and, as there is no need for further fermentation, the casks into which the wine is racked should be sulfured. This should be done with both red and white wines. The sulfur will remove a little of the color in the former case, but some of this will return as the sulfurous acid disappears, and the color remaining will be much more stable. Properly sulfured red wines will show more color at the end of six months than similar wines which have not been sulfured. It is neither necessary nor advisable to wait until the wine is perfectly clear before making the first racking. All wines made by commercial methods contain a few bacteria and many of them a great many, and even when wines taste dry there is often a trace of sugar left which is sufficient to serve as nourishment for the bacteria. It is verv desirable, therefore, to put the wine in such a condition that the bacteria will be deposited and the wine cleared as soon as possible. This is accomplished by racking without aeration, which eliminates the yeast and bacteria in the thick lees, and by sulfuring and cooling, which stop the action of the floating bacteria and cause them to be deposited. After the first racking, or before, if the wine is perfectly dry to the taste, the cellar should be kept as cool as possible and the wine pro- tected as much as possible from the air, in order to promote the settling of all the fine lees and hasten the perfect clearing of the wine. experiments: red wine. I. Over-ripe Grapes; Extraction by Heat; Use of Cooler; Pure Yeast. — On August 21st, at 3 p. m., about twelve tons of Zinfandel grapes were passed through the crusher and stemmer and pumped by means of a must pump into a vat furnished with a strainer. The grapes were very ripe, w r ith a large proportion of shriveled berries, and some quite dry. There were abundant signs of Mildew (Oidium) and some Black Mold (Aspergillus). The must showed 26% Balling and .55% of acid bul. 177—2 18 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. immediately after crushing and before the sugar from the dried grapes had diffused out into the must. There was a very little second crop present. Before placing the grapes in the vats, the latter were thoroughly washed and then swabbed with a 3% solution of sulfuric acid, which was left on the wood for several hours and then rinsed off with water. This treatment was given to all the vats used in the experiments. As soon as the vat was full, the must was drawn off and passed through the Gomot heater, from which it passed with a temperature of 140° F. back on to the skins in the original vat. This was continued with occasional stirring of the grapes until the contents of the vat had reached a temperature of 112° F. Before passing the must through the heater, 3.75 pounds of potassium meta-bisulfite were added to it to prevent the oxidizing effect of the air on the heated grapes. At 7 a. m. of the next day, August 22d, the must in the vat was red, but not deeply colored. The color was 2VR, 49.4. (See page 10.) The must at this time showed 27% Balling, owing to the diffusion of the sugar from the dried grapes into the must. The acid was .65% and the tannin .32%. The tannin was sufficiently high, but in order to increase the color it was decided to heat the vat again. At 9 a. m. the must was drawn off again and passed through the heater at 150° to 158° F. and back on to the skins until the whole vat showed 133° F. At 10 a. m. the color had increased to 2VR, 65.6, and the tannin to .352%. At 4 p. m. of the same day the red must was run through the cooler, reduced in temperature to 84° F., and a starter of 15 gallons of Cham- pagne yeast (see p. 13) added. Later, 250 gallons of water and 5 pounds of citric acid were added to the 1,700 gallons of must. The progress of the fermentation is shown below: Sugar. Temperature. Aug. 22, 6:00p. m 27% B. 84° F. Aug. 23, 6:00 a. m., fermenting well 25 85 12:00 m., water and acid added. 1:00p. m 22 84 Color, 3VR, 58; Acid, .66%; Tannin, .35%. 5:00 p. m., passed through cooler 19 90 7:00p.m.. 19 84 Aug. 24, 7:00 a. m., cooled to 84° F 16 92 Color, 1VR, 41.2. 12:00 m. 13i 87 5:00p. m 12 89 Aug. 25, 7:00 a. m., cooled to 90° 7 96 Color, 3VR, 27.5. 12:00m 5.2 90 6:00p. m 4.5 89 Aug. 26, 6:00 a. M - 2.5 93 12:00m.... 1.5 93 Racked into storage vat. Aug. 27, Nearly dry. A NEW METHOD OF MAKING DRY RED WINE. 19 On September 1st the wine showed 13.5% of alcohol and the color was 4VR, 26.3. On September 7th it was nearly clear, and was racked into puncheons, and in a few days was quite dry and clear. This experiment shows that heating the grapes to 112° F. is not suffi- cient to extract the maximum amount of color of ripe Zinfandel, even though the grapes are kept hot for fifteen hours. The fermenting wine was cooled three times. This was necessary because the regular cooling machine was not ready for use and a tem- porary and less efficient device had to be used. The fermenting wine was lowered 20° F. by the three coolings. This could have been done, as was proved later, by one cooling of two hours at the rate of 750 gallons per hour with the form of cooling machine finally adopted, or by two coolings of one hour at the same rate. The effect on the fer- FIG. 7. Color, Sugar, and Temperature Changes of Experiment I. mentation, however, was the same, and it continued without interruption until the wine was dry at six days, with 13.5% of alcohol. It is instruct- ive to compare this result with Experiment III, where the cooling was intentionally less complete. The wine of Experiment III required thirty days to become dry, though the alcohol in the wine was 1% less than in the wine of this experiment. An examination of Fig. 7 shows that the curve representing the diminution of color during fermentation is for the first three days very nearly parallel to that representing the diminution of the sugar-content. This probably indicates that the part of the color insoluble in alcohol is deposited in the same ratio as the alcohol is produced by the fermen- tation of the sugar. This rapid fall of the color ceases on the fourth day, the color which remains at this time being soluble in alcohol, and the further production of alcohol has little or no effect on it. The number 27 representing the color at the end of the fourth day probably 20 UNIVERSITY OF CALIFORNIA— EXPERIMENT STATION. indicates the whole amount of the original color in the grapes which is capable of remaining dissolved in the fermented wine. Whether all of this could have been extracted by the ordinary methods of fermentation is not shown by the experiment, but a comparison with the results of Experiment III indicates that it could not. The wine of the latter experiment, made from grapes similar to those used in Experiment I, had only 18 of color at the end of the fermentation. With regard to the stability of the color the comparison is even more strikingly in favor of the new method. At the end of three months the color of the wine of Experiment I was practically the same as when the fermentation had finished at four days, while the wine of Experiment III had lost 28% of its color. The broken line in Fig. 7 shows the temperature changes observed during the fermentation. A drop in temperature occurred four times. The first, a slight drop, was due to the addition of cold water used to dilute the must. The other three drops are the effects of the coolings which were given when the fermenting wine reached 90°, 92°, and 96°, respectively. A point of interest in this curve is that the abrupt changes in temperature had practically no effect on the rapidity of fer- mentation, as is shown by the even curve representing the disappear- ance of sugar. A slight irregularity of the sugar curve may be noticed opposite each jog representing a cooling on the temperature curve, but the irregularity is so slight that it is of no practical importance. This is shown equally in Experiment IX (see Fig. 10, page 25), where cool- ings of 13° F. and 14° F. show hardly any appreciable effect on the rate of fermentation. This lack of apparent influence of the cooling on the rate of fermentation is probably due to the fact that each cooling was accompanied by an aeration which invigorated the yeast and counter- acted the retarding effect of the lower temperature. II. Under-ripe Grapes; Use of Cooler ; Pure Yeast. — On August 25th about twelve tons of Zinfandel grapes were crushed, stemmed, and pumped into a fermenting vat, a starter of pure Champagne yeast being added as the grapes came into the vat. The grapes were poorly colored and not thoroughly mature. They showed 22.7% Balling and .66% of acid. The record of the fermentation is as follows: Sugar. Temperature. Aug. 25, 10:00 a. m . 22.7% B. 71° F. 7:00p. m 22.2 74 Aug. 26, 7:00 a. m., fermenting well 22.4 78 12:00m., fermenting well 20.4 78 7:00 p. m., fermenting well 19.2 88 Aug. 27, 7:00 a. m., cooled to 90° F 6.0 96 12:00 m 5.5 90 7:00p. M 4.0 92 Aug. 28, 7:00 v. m., racked into storage vat 1.0 94 Aug. 29, Dry. Sept. 7, The wine was still cloudy, but was racked into puncheons. A NEW METHOD OP MAKING DRY RED WINE. 21 This experiment illustrates the fact that even when the grapes are gathered incompletely mature, the temperature of the fermentation must be controlled in a hot climate if we are to produce a dry wine; and it also shows that we can not expect to produce wine of good quality from unripe grapes, as will be seen by comparing the quality of the various experiment wines as given on page 27. The checking of the fermentation at the time of cooling is more per- ceptible in this fermentation than in any of the others. In this case, as in that of the last cooling in Experiment IX, the checking at cooling is probably merely a coincidence and due in reality to the large amount of alcohol present at that stage of the fermentation. The upward course of the temperature curve (Fig. 8) during the first day without I SC. VtlU -Su^r %%.?/, / / s / y s yr 96 c '- A / \ / \ / \ / / ..J ZnJ. Dau Srd.D dv V 90 W J^th. J> \ / / $0°/ — , ** ' — »^_ ** ^.-^ N. n\ \ / ^^ *^ v -88° /\ / \ — / / / ^ / / P / _ / t. / ib / ( / f. — ~Temt>e rd ture 6v~^ V* O/o >■"■'■"■ ~~~~~ — _^__ ty' ht, Day 2. n J Ddtj 3rd J)dlj If.th.Day FKI. 9. Sugar and Temperature Changes of Experiment III. On September 1st the color was 5VR, 18.7, and the wine remained sweet until September 11th, when it was pumped over for an hour. On September 22d it was nearly dry, but still cloudy. This experiment shows the danger of allowing the fermentation to rise as high as 98° F., but also the possibility of getting it through com- pletely within a few weeks without any serious damage to the quality. In this case the chief loss was the extra trouble and time needed to get the wine dry. The cooling ought to have been done very early on the morning of August 27th instead of waiting until the afternoon. It was a mistake also to add the sulfite at noon on August 27th. The sulfite, if used at all, should have been added on the evening of August 26th, before the high temperature was reached. The curves of Fig. 9 are instructive in several particulars. Compared with those of Fig. 10 they show plainly that the hottest fermentation A NEW METHOD OP MAKING DRY RED WINE. 23 is not necessarily that which finishes first. In Fig. 9 the sugar curve drops very rapidly from the middle of the first day to the middle of the second, when it comes to an abrupt stop, becoming almost hori- zontal for twenty-four hours. The rapid loss of sugar is accompanied by a corresponding rapid rise in temperature due to the former. The temperature of 98° reached practically stops the fermentation com- pletely for twenty-four hours. This is shown both by the sugar remain- ing unchanged and by the temperature falling. The wine was cooled from 98° to 90° by the machine, and then instead of rising again, as it would have done if there had been any active fermentation going on and as it did after every cooling in Experiments I and IX (see Figs. 5, 7, and 10), it fell to 88°. This shows that there was not enough fermenta- tion going on to counterbalance the heat lost by radiation. The vat had "stuck." The "sticking," however, was not so serious as it often is in vats which rise higher than 98°, or stay at this temperature for some time. The cooling as soon as the temperature reached 98 n , and the aeration practiced afterwards, revived the fermentation, as is shown by the gradual rise of the temperature curve and the fall of the sugar curve. IV. Over-ripe Grapes; Dilution; Addition of Acid. — On September 1st at 11 a. m., five tons of Zinfandel grapes were crushed and stemmed into a fermenting vat. The grapes were over-ripe, some of the berries dried, and there was a large amount of black mold. The must showed 29.5% Balling and .7% acidity. It was diluted with 250 gallons of water, to which was added 6 pounds of tartaric acid. The temperature of the grapes at the start was 73° F., and within thirty-six hours had risen to 90° F. at the bottom of the vat and 104° F. in the cap. The fermentation stuck when the sugar had fallen to 7% Balling, and could not be revived by cooling. The wine was used for port, as an attempt to make it dry would probably have resulted in spoiling it completely. This experiment shows the futility of attempting to make a dry wine in a hot climate from over-ripe grapes, even if the acidity is increased artificially and the sugar decreased by dilution, unless we adopt some means of controlling the temperature. V. Over-ripe Grapes; Dilution; Addition of Acid; Control of Tem- perature by Use of Sulfites. — A vat of Zinfandel grapes, similar in every respect to those used in Experiment IV, was diluted and acidified in exactly the same way and an attempt made to control the temperature by the occasional addition of a calculated amount of a solution of potassium meta-bisulfite. The experiment was a failure, as the temper- ature ran up to a higher point than that reached in Experiment IV, and the wine stuck with 12% Balling. 24 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. This was by no means a conclusive test of the method, as the amount of sulfite used was too small. The use of sulfite was tested again in Experiment VII. VII. Ripe Grapes; PureYeast; Control of Temperature }>y Use of Sulfites; Pomace and Must Fermented Separately. — (Debono method, see Bulletin No. 167, p. 18.) The grapes used in this experiment con- sisted of 29% Charbono, 29% Lenoir, and 42% Burger; nine tons in all. They were ripe and in good condition, the mixture showing 22.5% Ball- ing. The following is the record of the fermentation: September 2d, 3:00 p. m. Crushed and added potassium meta-bi- sulfite solution (=.06 per mil.) and 15 gallons of Champagne yeast accustomed to sulfite. September 3d, 7:00 a. m. The fermentation had commenced, and the must was drawn off into an open vat and given another dose of sulfite solution twice as large as the first dose (=.12 per mil.). The must and pomace were then allowed to ferment separately. The further progress of the fermentation is shown by the following table: Must. Temp. Sept, 3, 6:00 p. m Sept, 4,7:00 a. m. 5:00p. m . ._. Added more sulfite (=.12 per mil.). Sept, 5, 7:00 a.m. 12:00 m 74° F. 86 94 96 98 Sugar. 22.5% B, 13 10 4.5 2.5 Pomace. Temp. 74° F. 94 98 100 100 Sugar. 22.5% B. 6 2.5 1.5 1.0 :00 p. m. Sept. 6, Sept. Sept. Sept. The must, which had fallen to 2% Balling, was now pumped back on to the pomace and allowed to stand, after a thorough stirring, for twenty-four hours. When first mixed, the vat showed a temperature of 98°'F. and 2% Balling. 7:00 a. m. Temperature 92° F., sugar 1.5% Balling. 12:00 m. Racked off pomace into open vat and aerated. The color of the wine was 2VR, 20.3. 7. The wine was transferred to a storage cask. 11. The wine still tastes a little sweet and was pumped over for one hour. 22. Still shows a slight trace of sweetness. While this wine did not stick entirely and finally went through per- fectly dry, there is no evidence in the experiment that the addition of the sulfite had the slightest effect in preventing high temperatures. The grapes had only 22.5% of sugar and were cool (74° F.) when crushed. If we reckon that each per cent of sugar in fermenting heats the must 1.17° F.,* the temperature of the fermenting grapes should have reached 97.5° F. when 20% of sugar had fermented out. This corresponds almost exactly to what occurred, so that there is no delay of fermenta- tion or moderation of temperature that can be accredited to the sulfite. If sulfites are to be effective for this purpose they must evidently be used in larger doses than those adopted in this experiment. *See Bulletin No. 174, pp. 17, 18, 19. A NEW METHOD OP MAKING DRY RED WINE. 25 IX. Ripe Graces; Pure Yeast; Extraction by Heat; Temperature Controlled by Cooling Machine. — The grapes used in this experiment were a mixture of Grenache 45%, Zinfandel 31-J%, and Lenoir 23-J%. They were mature and in good condition except for a few dried and moldy grapes and some green bunches of second crop in the Zinfandel. The must was passed through the heater and back on to the pomace until the temperature of the whole was 125° F., and the whole well stirred. This was on the afternoon of September 8th, and at 9 a. m. of September 9th the must was drawn off the pomace and cooled. The FIG 10. Sugar and Temperature Changes of Experiment IX. color of the must was 1 VR, 53.4. Yeast was added, and the fermenta- tion proceeded as follows: Sept. Sugar. Temperature. 9, 4:O0p. m 23.0%B. 84° F. 7:00 p. m., cooled to 80° 21.5 87 10:00 p. m 19.0 80 Sept. 10, 7:00 a. m , cooled to 79° 12.5 92 10:00 a.m. 12.0 79 10:00p.m. 8.0 85 Sept. 11, 7:00 a. m., cooled to 78° 4.5 92 12:00m 3.0 78 6:00 p. m 2.5 78 Sept. 12, 7:00 a. m., transferred to a storage cask 1.0 79 Sept. 19, Still a little sweet and still fermenting. Sept. 23, Pumped over for one hour. This experiment exemplifies the benefit to be derived from complete control of the temperature by efficient cooling. The wine required a month to become quite dry, but at the end of that time all the sugar was eliminated, although the wine contained 15.1% of alcohol. It became dry as soon as did the wine of Experiment III, which contained only 12.5% of alcohol but was allowed to reach a temperature of 98° F. The wine was cooled three times, a total of 27° F. being taken out by the cooler. The same result could have been obtained with considerably 26 UNIVERSITY OP CALIFORNIA — EXPERIMENT STATION. less cooling if in order to avoid night work the first cooling had not been done sooner than was necessary. The last cooling, moreover, was much greater than was necessary and might have been dispensed with altogether. The sugar curve in Fig. 10 is in marked contrast with that of Fig. 9. The drop in the curve is just as rapid during the first thirty-six hours, but instead of being checked there it continues for forty-eight hours until the sugar reaches 3%. Here the fermentation is checked slightly but not " stuck," and the wine was practically dry in three days when Experiment III still had 5% of sugar. This difference is undoubtedly due altogether to the fact that Experiment IX was not allowed to rise above 92° F. experiments: white wine. VI. Defecation by Sulfuring; Pure Yeast; Fermentation in Punch- eons. — The grapes used in this experiment were about eight tons of Johannisberg and Franken Riesling. They were in excellent condition and were a striking example of what can be done by proper care and cultivation, even with varieties presumably so unsuited to a hot climate as the grapes of the Rhine. The grapes were thoroughly and evenly matured and showed no sunburn, mildew, or mold. They were crushed into an open vat, and the first 500 gallons of must which ran off was pumped into a heavily sulfured cask. After settling for twenty-four hours, the must was drawn off into three sulfured puncheons and started with pure Champagne yeast. The following is the record of the fermentation: Date. Sugar. Temperature. August 26th 22.5% B. 78° F. August 27th 16.7 85 August 28th 9.1 88 August 29th 4.7 88 August 30th 2.8 88 August 31st 2.0 87 The wine fermented slowly, and on September 2d was still a little sweet. It was then pumped-over to aerate it, and on September 8th was dry and nearly clear. Via. Fermentation Started on Skins; Pure Yeast; Fermentation in Puncheons. — After drawing 500 gallons of must off the crushed grapes of Experiment VI, the remainder was left on the skins for twelve hours. At the end of this time a slight fermentation was perceptible, and the must was drawn off into sulfured puncheons and started with Cham- pagne yeast. The folloAving is the record of the fermentation: Date. Sugar. Temperature. August 26th 25.0% B. 78° F. August 27th 8.7 92 August 28th 5.9 89 August 29th 4.4 87 August 30th 4.0 86 August 31st 4.0 84 A NEW METHOD OP MAKING DRY RED WINE. 27 The wine was treated in the same way as in Experiment VI, but on September 8th it was still cloudy and slightly sweet. On September 22d it was cloudy and fermenting and had a slight brownish color. It finally became nearly dry about October 21st, and the brownish tint had almost disappeared, though it was still cloudy. A 50-gallon cask of each experiment wine was shipped to Berkeley and stored in the Station cellar. They were tasted on November 24, 1905, with the following results: Exp. I. Perfectly clear, full flavor, sound, fruity, and agreeable, but with a slight taste of dried grapes. Exp. II. Cloudy, of poor color and a little mousey. Exp. III. Clear and a good wine, but inferior to Experiment I. Exp. VII. Clear, odor peculiar, flavor good. Exp. IX. Clear, a little better than Experiment VII. Exp. VI and Via. Nearly clear, contain a small amount of sugar. Good, full-bodied wines of somewhat Sauterne type. Exp. VIII. Tastes flat, but otherwise a fair wine of neutral Sauterne type. Order of merit of the red wines: I, III, IX, VII, II; of the white: VI, Via, VIII. The analyses of the wines made on December 1, 1905, by Prof. George E. Colby, are given in the following table: TABLE II. Analyses of Experiment Wines from Fresno. December 1, 1905. Exp. I. Exp. II. Exp. III. Exp. VI Exp. Via. Exp. VII. Exp. IX. Exp. VIII. Specific gravity... Alcohol, per cent by volatile Alcohol, grams per 100 cc "Extract," grams per 100 cc Ash, grams per 100 cc Acidity, total, as tartaric; grams per 100 cc Acidity, fixed, as tartaric; grams per 100 cc Acidity, volatile, as acetic; grams per 100 cc Sugar, total, grams per 100 cc Tannin .9935 13.80 10.95 3.02 .37 .457 .324 .108 .292 .9955 12.30 9.7G 3.10 .38 .397 .285 .090 .200 .160 .9970 12.50 9.92 3.25 .45 .442 .300 .114 .300 ,240 .9900 13.25 10.51 2.19 .25 .390 .300 .072 .9925 14.50 11.51 2.80 .37 .337 .292 .036 .300 .9980 11.70 9 29 3.30 .57 .375 .300 .060 .300 .280 .9930 15.10 11.98 2.60 .41 .315 .232 .066 .9910 12.60 10.00 2.07 .33 .375 .315 .048 .260 A puncheon of each of the experiment wines was kept separate and left at the cellar where they were made. They were tasted on January 1, 1906. The wines made by the new method, I and IX, were bright, dry, clean-tasting, and of good flavor. The wines made in the ordinary way, III and VII, were dry and clean-tasting, clear but not bright. The main differences to be perceived in the wines were that those made by the new method were brighter and more astringent than the others. The white wines were all in excellent condition, and, on the whole, surprisingly good wines for the district and the conditions in which they were made. The Burger was rather flat and lacking in acid. 28 UNIVERSITY OF CALIFORNIA — EXPERIMENT STATION. COLOR-CHANGES IN THE RED WINES. The record of the color-changes in the various red wines is shown in the following table. The same record is shown graphically in Pig. 11. TABLE III. Color of Experiment Wines. Exp. I. Exp. II. Before fermentation '• 2VR,65.6 After fermentation | 3VR, 27.5 R,13.1 3 months' fermentation : 3R+Y, 7.5 4i months' fermentation 3VR, 16.2 3R+Y, 5.3 Exp. III. Exp. VII. Loss of color in 3 months. Loss of color in Ah months 41°/ 43% 60 I 5VR, 18.7 5VR, 13.4 R, 10.2 28% 45/ COLOR Remaining